Table 12. Mean and Standard Deviation of the Performance of Different Methods for the Models Built on the Filtered Data Sets for Nonspecific and Cytotoxic Compounds for TPO and TR.

end point	method	F1 score	MCC	balanced accuracy	AUC
TPO	RF	0.81 (± 0.05)	0.63 (± 0.10)	0.78 (± 0.04)	0.91 (±0.03)
	LR	0.79 (± 0.07)	0.59 (± 0.13)	0.79 (± 0.06)	0.87 (± 0.04)
	XGB	0.80 (± 0.04)	0.60 (± 0.09)	0.79 (± 0.04)	0.89 (± 0.02)
	SVM	0.79 (± 0.03)	0.58 (± 0.07)	0.78 (± 0.04)	0.89 (± 0.03)
	NN	0.79 (± 0.05)	0.58 (± 0.11)	0.77 (± 0.05)	0.88 (± 0.04)
TR	RF	0.68 (± 0.05)	0.39 (± 0.10)	0.65 (± 0.04)	0.88 (± 0.02)
	LR	0.63 (± 0.03)	0.28 (± 0.05)	0.62 (± 0.02)	0.77 (± 0.05)
	XGB	0.67 (± 0.05)	0.37 (± 0.10)	0.64 (± 0.04)	0.85 (± 0.04)
	SVM	0.66 (± 0.04)	0.34 (± 0.07)	0.66 (± 0.05)	0.82 (± 0.04)
	NN	0.64 (± 0.04)	0.29 (± 0.07)	0.61 (± 0.04)	0.81 (± 0.04)